Studies on the Biosynthesis of Porphyrin and Bacteriochlorophyll by Rhodopseudomonas Spheroides 5

Biochem. J. (1964) 90, 607 607

Studies on the Biosynthesis of Porphyrin and Bacteriochlorophyll by Rhodopseudomonas spheroides 5. ZINO-PROTOPORPHYRIN CHELATASE*

BY A. NEUBERGER AND G. H. TAIT Department of Chemical Pathology and M.R.C. Research Group in Enzymology, St Mary's Hospital Medical School, London, W. 2 (Received 31 July 1963) Of the metal complexes of protoporphyrin which are of biological importance those of magnesium MATERIALS AND METHODS and ferrous iron are of special importance. Mag- Chemica,1. [65Zn]Zinc chloride was obtained from The nesium protoporphyrin is most probably an inter- Radiochemical Centre, Amersham, Bucks. Tween 20 and mediate in the biosynthesis of chlorophyll (Granick, Tween 80 were gifts from Honeywell and Stein Ltd., 1961) and of bacteriochlorophyll (Gibson, Neu- London. De-ionized water, which was used throughout, berger & Tait, 1963), while ferrous protoporphyrin was prepared from distilled water by passing it through a (protohaem) is a component of haemoglobin, myo- column of Bio-Deminrolit (The Permutit Co. Ltd., Lon- don). globin, cytochromes and other haem proteins. The Ether was dried over anhydrous Na2SO4 and redistilled enzymic formation of magnesium protoporphyrin before use. has not been described, but that ofprotohaem from All other chemicals were obtained or prepared as Fe2+ ions and protoporphyrin is well documented. described by Gibson et al. (1963). Ferrochelatases have been found in mitochondria Organisms. The strain of Rhodopseudomonas spheroides from rat liver (Labbe & Hubbard, 1961) and pig and the conditions of growth have been described by liver (Porra & Jones, 1963 a, b) and in haemolysates Gibson et al. (1963). Except where stated otherwise the from duck blood (Yoneyama, Oyama, Sugita & organisms used to make enzyme extracts were grown semi- Yoshikaya, 1962) and chicken blood (Schwartz, anaerobically in the light as described by Gibson, Neu- & 1959). These enzymes berger & Tait (1962). For certain experiments organisms Hill, Cartwright Wintrobe, were grown aerobically in the dark. 'High' or 'low' con- have been partially purified and their specificities tent of 02 obtained by varying the extent of aeration gave towards various porphyrins and metal ions have unpigmented or pigmented organisms respectively (cf. been tested. They act on all porphyrins with two Lascelles, 1959). In the experiments described here the carboxyl groups. They are, however, more specific content of 02 was not measured, but this factor was towards the metal; the enzyme in duck blood controlled empirically by observing the extent of pig- (Oyama, Sugita, Yoneyama & Yoshikaya, 1961) mentation of the organisms. uses Zn2+ ions at 10 % and Co2+ ions at 2 % the rate of Fe2+ ions, the one from rat-liver mito- Enzyme sources chondria (Labbe & Hubbard, 1961) uses C02+ ions Chrornatophores. These were prepared according to more efficiently than the one in duck blood but is Gibson et al. (1963) from R. spheroides. In all experiments completely inactive with Zn2+ ions; it also uses except where stated otherwise the chromatophores were Mn2+ ions to a small extent. prepared from organisms grown semi-anaerobically in the During the course of unsuccessful experiments light in the malate medium of Lascelles (1956). designed to detect a magnesium-incorporating en- Protoptasts. These were prepared as described by Karun- zyme in chromatophores from Rhodopseudomonas airatnum, Spizizen & Gest (1958). They were lysed by spheroides the production of zinc protoporphyrin dilution of the suspension in 20% sucrose with a large volume of water. The lysed particles were collected by was noted, and the rate of its formation was greatly centrifuging, washed with 0 05M-potassium phosphate, stimulated by bringing about the reaction in a pH 7 4, and resuspended in the same buffer to a concen- water-ether emulsion. Under these conditions a tration of 10-30 mg. of protein/ml. and kept frozen at marked synthesis of zinc protoporphyrin was also -200. detected with mitochondria from a number ofmam- Mitochondria. These were prepared from guinea-pig malian tissues. The properties of these zinc-proto- liver and from rabbit liver, heart and kidney as de- porphyrin chelatases, particularly the one in R. scribed by Hogeboom (1955). After being washed twice spheroide8, are described in detail and their possible in 0*25M-sucrose the mitochondrial pellet was suspended metabolic functions discussed. in 0*05M-potassium phosphate, pH 7*4, to a concentration of 10-20 mg. of protein/ml. and kept frozen at * Part 4: Gibson, Neuberger & Tait (1963). -200. 608 A. NEUBERGER AND G. H. TAIT 1964 Estimations total) in the final ether solution the activity is expressed in terms of the theoretical amount of zinc protoporphyrin Protein. The protein content of chromatophores, proto- formed in the assay. plasts and mitochondria was determined by the method of Ferrochelatase. Tris buffer, pH 8-4 (150 ,umoles), proto- Lowry, Rosebrough, Farr & Randall (1951), with crystal- porphyrin (50 ,um-moles), FeSO4 (50 ,um-moles), ascorbic line bovine plasma albumin (Armour and Co. Ltd.) as acid or GSH (10-20 Mmoles) and enzyme extract (0 5- standard. 2-0 mg. of protein) were incubated in a total volume of Porphyrin8. The concentration of porphyrins and of 0 5 ml. under N2 for 2 hr. at 37°. The treatment of the metaUoporphyrins was determined by measuring the ex- mixture followed exactly that described for the assay of tinction of solutions in ether at the absorption maximum zin¢-protoporphyrin chelatase. Since haemin is insoluble in the Soret band, and by using the appropriate molar in ether the extent of the reaction was taken as the differ- extinction coefficient. ence between the amounts of protoporphyrin recovered in Radioactivity. Solutions were plated on 6-25 Cm.2 the final ether solution when experiments were done in the aluminium planchets, dried in vacuo and counted at in- presence and absence of FeSO4 respectively. finite thinness in a Nuclear-Chicago gas-flow counter with a Micromil end window and operated at the centre of the plateau. One Mco of 65Zn gave 24400 counts/min. under RESULTS these conditions. Alternatively, solutions (0 5 ml.) were mixed with 5 ml. of a dioxan-based scintillator (Bray, Preliminary observations 1960). Radioactivity was measured in a I.D.L. scintillation The finding that the enzymic conversion of proto- counter with the upper-gate-discriminator bias set at 50 v. porphyrin, but not of magnesium protoporphyrin, Under these conditions 1 MC of 65Zn gave 253800 counts/ into the corresponding ester was min. monomethyl inhibited EDTA that Radioactive spots on paper chromatograms were located strongly by suggested proto- by radioautography. porphyrin had first to be converted into a metallo- Infrared-absorption spectra. Porphyrin (50-100 Mg.) in a porphyrin before it could be methylated (Gibson et smal volume of ether was mixed with 6 mg. of powdered al. 1963). That the insertion of the metal was KCl and the suspension was evaporated to dryness. The catalysed by an enzyme was indicated by the fol- resulting powder was pressed to form a micro-disk. The lowing experimental results. Protoporphyrin and infrared spectrum was measured on a Unicam SP. 200 chromatophores in tris buffer, pH 8f4, were in- spectrophotometer. cubated at 370 for periods up to 2 hr., after which EDTA was added, followed by S-adenosyl[Me-14C]- Assay of enzyme activities methionine. The mixture was incubated for a Zinc-protoporphyrin chelatase. To incubation mixtures further 2 hr. The amount of radioactivity incor- (vol. 0 5 ml.; cf. legends to Figs. 1-3 and Table 1) was porated into porphyrin was proportional to the added 4 ml. of acetone-0- N-NH3 (9: 1, v/v). After centri- duration of the first incubation in the absence of fuging, the supernatant was extracted twice with 3 ml. of EDTA. The formation of metalloporphyrin was also light petroleum (b.p. 60-80°). This removes the bacterio- the of radio- chlorophyll and carotenoids of the chromatophores. To the pH-dependent; greatest incorporation activity into porphyrin took place when the chro- aqueous phase, containing the porphyrins, were added 0-3 ml. of saturated NaCl, 0-5 ml. of 0-5M-KH2PO4 and matophores and protoporphyrin were incubated at 4 ml. of ether. After thorough mixing the ether layer was pH 8S4 before addition of EDTA. A large-scale removed and analysed for porphyrin and zinc porphyrin. experiment was performed in which protoporphyrin, This extraction method is modelled on that of Granick S-adenosyl[Me-14C]methionine and chromatophores (1961). In assays with 65ZnC12 this ether solution was from R. spheroides, and tris buffer, pH 8*4, were washed with 3 ml. of water before portions of it were taken incubated for 4 hr. After incubation, the por- for measurement of radioactivity. Most of the assays were phyrin mixture was extracted and purified by the done with protophorphyrin as substrate and therefore the method of Granick (1961), which does not hydrolyse extinction of the was at ether solution measured 404 m,u acid-labile This mix- and 415 mMu, which are the absorption maxima of proto- metalloporphyrins. porphyrin porphyrin and zinc protoporphyrin respectively. When ture was chromatographed on paper in the luti- other porphyrins were used as substrates the wavelengths dine-ammonia system of Granick (1961). All the at which readings were measured were chosen accordingly. radioactivity was located in an area corresponding From the known molar extinction coefficients in ether of to a porphyrin with one free carboxyl group. This protophorphyrin (404 mM, 1-35 x 105; 415 mM, 0 94 x 105) portion of the paper was cut out and the eluted and zinc protoporphyrin (404 mp, 0-75 x 105; 415 mrI, porphyrin was identified as a metalloporphyrin on 2-40 x 105) the concentration of each component was calcu- the basis of its spectrum in ethereal solution which, lated in a manner similar to that described by Dresel & in the visible region, had peaks of almost equal Falk (1956). It was assumed that protoporphyrin zinc and intensity at about 545 mM and 590 m,. Zinc, mag- protoporphyrin behave in identical fashion during the extraction and that the relative amounts are the same in the nesium and calcium protoporphyrins all have final ether solution as they are at the end of the experiment similar spectrum bands but there are slight differ- in the incubation mixture. To correct for slight differences ences in the intensities of the two peaks. Probably in the recovery of porphyrins (usually about 60% of the owing to the presence of some impurity the nature Vol. 90 ZINC-PROTOPORPHYRIN CHELATASE 609 of the experimental metalloporphyrin could not be either in the presence or absence of chromato- deduced. phores: Cu2+, Co2+, Al"+, Mn2+, Fe2+, Fe3+, Ca2+, It was found that formation of metalloporphyrin Ba2+, Sr2+, Mg2+. Even though the spectra of some could be estimated directly, without coupling it to of the expected metalloprotoporphyrins are not the methylation, by measuring the extinction of the known, any formation would be expected to change extracted porphyrins at two different wavelengths the ratio of extinctions at 404 m,u and 415 m,u from in the Soret region of the spectrum (see the Materials and Methods Section). For metalloporphyrin formation only tris buffer, protoporphyrin 10 and chromatophores were required and it was con- sidered that Mg2+ ions, or some magnesium complex present in the chromatophores, might be the 00 metal substrate. It was subsequently found that the metal ion was present in the tris buffer. With Go- C6 I fresh tris buffer made up in de-ionized water no -4 metalloporphyrin was formed unless Zn2+ ions were .00 added; moreover, no other metal ions could replace 0 Zn2+ ions. Therefore in all further experiments on 0 'o the enzymic formation of zinc protoporphyrin suit- ;0l 4 able controls were included to check that con- 0 tamination of the reagents with Zn2+ ions was N small. It was not found necessary to take any pre- .6 cautions other than using de-ionized water for making up all solutions. 0-4 0-8 1-2 1-6 2-0 Chromatophores (mg. of protein) Zinc-protoporphyrin chelatase in chromatophores Fig. 1. Formation of zinc protoporphyrin in the presence The formation of zinc protoporphyrin in the pre- ofdifferent amounts ofchromatophores. Tris buffer, pH 8-4 sence of increasing amounts of chromatophores is (150 ,moles), protoporphyrin (34 ,m-moles), ZnCl2 (20 ,um- shown in Fig. 1. The small but significant formation moles) and chromatophores (as stated) in a total volume of of zinc protoporphyrin in the absence of chromato- 0-5 ml. were incubated at 370 for 1 hr. The porphyrins were phores was no greater than was obtained with isolated and the amount of zinc protoporphyrin formed was chromatophores previously boiled for 3 min. or with estimated as described in the Materials and Methods the supernatant obtained on centrifuging crushed section. organisms at 105 000g for 90 min. The progress of the reaction was approximately linear for 60 min., when using up to 1 1 mg. of protein/assay. The - 9 2 reaction was carried out at temperatures from 50 -cm 8 to 380, and the heat of activation was calculated to g 7 be 14800 kcal./mole. About 70 % of the activity cB0O was lost by heating the chromatophores in tris .6 buffer, pH 8-4, for 5 min. at 55°. This loss in 5 activity could be abolished by heating the chroma- 01 tophores at 55° in the presence of protoporphyrin .R4 (50 ,um-moles/0.5 ml.), but not in the presence of PLI zinc chloride (15 ,um-moles/0-5 ml.). The formation of zinc protoporphyrin in the pre- 0-+ 2 sence of increasing amounts of Zn2+ ions is shown in Fig. 2. There is a significant formation, about .5 20 % of the maximum value, in the absence of 0 2 4 6 8 10 12 14 16 18 20 added Zn2+ ions, and with low concentrations of ZnCl2 (,um-moles) added Zn2+ ions more than the theoretical amount ofzinc protoporphyrin is formed. The Km for proto- Fig. 2. Formation of zinc protoporphyrin in the presence of was and that for zinc different amounts ofZnCl2. Tris buffer, pH 8-4 (150 ,umoles), porphyrin 200 pM, chloride, protoporphyrin (39 ,um-moles), ZnCl2 (as stated) and chro- which cannot be calculated accurately because of matophores (0-97mg. of protein) in a total volume of 0-5ml. the contamination, is probably about 40-S50 M. were incubated at 370 for 1 hr. The porphyrins were Under the conditions of assay none of the following isolated and the amount of zinc protoporphyrin formed was ions (tested at 100 ,um-moles/assay) gave rise to estimated as described in the Materials and Methods significant formation of metalloprotoporphyrins section. 39 Bioch. 1964. 90 610 A. NEUBERGER AND G. H. TAIT 1964 that obtained with protoporphyrin itself. The same 7 ions were also tested, at 100 ,um-moles/assay, in the 0 6 presence of 10 1am-moles of zinc chloride. The Co2+ .z - salt inhibited zinc protoporphyrin formation by 5 l?P4 . 70 and the Mn2+ salt by 50 other cations were F-4 A % %; 0 -- P, an 4 apparently without effect (but see below for experi- 0 -+D 0 3 ments with [65Zn]zinc chloride). A number of k 0 metal at P4 i agents known to form complexes tested 0 2 0 03 mM (diphenylthiocarbazone, 1,10-phenanthro- - 4 - line and 8-hydroxyquinoline) were inactive, but N EDTA at 30 ,um-moles/assay inhibited by 60 % the LA formation of zinc protoporphyrin promoted by 7-0 7-3 7-6 7 9 8-2 8-5 8-8 10 pm-moles of zinc chloride. pH The pH-activity curve in tris buffer (Fig. 3) Fig. 3. pH-activity curve for zinc-protoporphyrin chelat- shows an optimum in the region of pH 8. The ase. Tris of pH stated (150 umoles), protoporphyrin activity was the same in triethanolamine of the (34 gm-moles), ZnC12 (10 ,m-moles) and chromatophores same pH, but was considerably lower in phosphate, (0 75 mg. of protein) in a total volume of 0-5 ml. were in- probably owing to formation of zinc phosphate. cubated at 370 for 1 hr. The porphyrins were isolated and The effects ofascorbic acid and GSH were examined, the amount of zinc protoporphyrin formed was estimated since these reducing agents are necessary for de- as described in the Materials and Methods section. monstrating ferrochelatase activity (Labbe & Hub- bard, 1961). GSH markedly inhibited the reaction, but ascorbic acid at concentrations up to 20 ,moles/ assay had no effect. Protoporphyrinogen could be used as substrate instead of protoporphyrin: however, the former was completely inactive in the presence of ascorbic acid. This is probably because ascorbic acid inhibits the oxidation of protoporphyrinogen to the true substrate protoporphyrin. E When zinc protoporphyrin is formed enzymically there is a change in the spectrum in the visible region between 490 mp and 630 mp (Fig. 4). The spectrum of the final ether solution (see the Materials and Methods section) obtained from incubations containing protoporphyrin, Zn2+ ions and chromatophores is intermediate between that of protoporphyrin and zinc protoporphyrin. The for- 520 560 600 640 mation of zinc protoporphyrin was further con- 500 540 580 620 firmed in experiments with 6Zn2+ ions, where the Wavelength (mu) increase in the ether-soluble radioactivity paralleled Fig. 4. Spectra in ether solution of protoporphyrin (-----) the formation of zinc protoporphyrin as measured zinc protoporphyrin (. ) and porphyrins extracted after by extinction change (Table 1). The non-enzymic incubating protoporphyrin, ZnCl2 and chromatophores formation of zinc protoporphyrin noted above ( ), Table 1. Assay of zinc-protoporphyrin chelatase in the presence of [65Zn]zino chloride Tris buffer, pH 8 4 (100,moles), protoporphyrin (30,m-moles), 65ZnC12 (15-6 pm-moles; 0 04,uc/,um-mole; 980 counts/min./,um-mole at infinite thinness on gas-flow counter) and chromatophores (3.4 mg. of protein) in a total volume of 0 5 ml. were incubated for the times stated at 37°. The porphyrins were isolated and the amount of zinc protoporphyrin and the radioactivity in the final ether solution were estimated as described in the Materials and Methods section. zmc_-. protoporpiyrm7;n t+h; u;ounts/mm./jAm-moie^~~~~~~uns/mm./1; -1 1 Time of incubation (,um-moles/ml. Counts/min./ml. of zinc (min.) of ether) of ether protoporphyrin 0 0-07 4 55 25 0-68 256 392 50 1-01 416 414 75 143 616 432 100 1.50 608 405 100 (no chromatophores) 3.53 272 518 100 (no protoporphyrin) ).,0 15 Vol. 90 ZINC-PROTOPORPHYRIN CHELATASE 611 (Fig. 1) was also confirmed. That the ether-soluble stimulation was two- to three-fold. Butanol, ace- zinc is all in the form of zinc protoporphyrin was tone, ethylene glycol, propylene glycol (all 0-2 ml.) shown by chromatography followed by radioauto- were inactive, as was deoxycholate (0.5 %, w/v). In graphy (see below). It may also be noted that the the presence of ether, as in its absence, the incor- molar radioactivity of the zinc protoporphyrin is poration of 65Zn2+ ions followed closely the forma- lower than expected. tion of zinc protoporphyrin as measured spectrophotometrically (Fig. 5). That the ether-soluble Assay of zinc-protoporphyrin chelatase f5Zn is actually in the form of zinc protoporphyrin in the presence of ether was shown as follows. An incubation was done Under the assay conditions described above the with protoporphyrin, 65Zn2+ ions and chromato- maximum conversion of protoporphyrin into zinc phores in the presence of ether, and, as judged from protoporphyrin was about 30 %. No magnesium the visible spectrum, 80 % of the porphyrin was in protoporphyrin formation was detected and it was the form of the zinc complex. The porphyrin was thought that magnesium in the ionic form might purified and chromatographed on paper in luti- first have to be transferred to a lipid phase before dine-0-05N-ammonia (10: 7, v/v) (Granick, 1961). being incorporated into protoporphyrin. Therefore It ran as a single spot, having an RF identical with the reaction was carried out in the presence ofether. that of zinc protoporphyrin and protoporphyrin To the 0-5 ml. of assay system was added 0-2 ml. of (these two compounds not being separated by this ether. The tube was stoppered, shaken vigorously solvent) and on radioautography all the radio- to obtain an emulsion and incubated. From time to activity was associated with this spot. time the tube was shaken to keep the emulsion In Table 1 and Fig. 5 it can be seen that the intact. No formation ofmagnesium protoporphyrin molar radioactivity of the isolated zinc protopor- was detected but when these conditions were tried phyrin is considerably lower than that of the with Zn2+ ions instead of Mg2+ ions a large stimu- [65Zn]zinc chloride added. There are a number of lation of zinc protoporphyrin formation was found possible reasons for this, but the most obvious one (Fig. 5). Ether had no effect on the non-enzymic is the presence of non-radioactive Zn2+ ions as a formation of zinc protoporphyrin but consistently contaminant of the reagents. This explanation stimulated the enzymic reaction by a factor of five agrees with the findings discussed above that a to six. With methanol (0-2 ml.) or Tween 80, at a small quantity of zinc protoporphyrin is formed in final concentration of 0-5 %. instead of ether the the absence of added Zn2+ ions and that when small amounts of Zn2+ ions are added the formation of zinc protoporphyrin is often higher than the 1600 theoretical amount (cf. Fig. 2). An experiment in the presence of ether was performed under the con- 1200 E ditions described in Fig. 5 but with quantities of =4 [66Zn]zinc chloride varying from 8 to 100 ,m-moles. 6 X By comparing the molar radioactivities of the zinc P4a' chloride and the zinc protoporphyrin, and by the use of values obtained on the gas-flow counter and -4 the scintillation it was calculated that 0 counter, 400 P.o there were about 20 ,mg. ions ofZn2+ contaminating -4d6- 0 each 0-5 ml. of assay mixture. However, from the low blanks obtained in the absence of added Zn2+ ions (cf. Fig. 2) all the contaminating zinc does not 0 15 30 45 60 appear to be 'free' to form zinc protoporphyrin Time of incubation (min.) unless additional Zn2+ ions are added. For in- Fig. 5. Effect of ether on zinc-protoporphyrin-chelatase stance, in an experiment with ether only 3-6 ,m- activity. Tris buffer, pH 8-4 (100 ,umoles), protoporphyrin moles of zinc protoporphyrin were formed in the (39 ,m-moles), ZnC12 (30 ,um-moles; 0-02 ,uc/,m-mole; absence of added Zn2+ ions, whereas 20 pm-moles 429 counts/min./,um-mole at infinite thinnesss) and chro- were formed on the addition of 10 umg. ions. On matophores (1-4mg. ofprotein) were mixed in a total volume the other hand the low molar radioactivity of the of 0-5 ml. To one set oftubes 0-2 ml. ofether was added, the zinc protoporphyrin may be due to the fact that all tubes were stoppered and the contents mixed thoroughly. the metalloporphyrin formed is not zinc proto- All tubes were incubated at 370 for the times stated. The or all in the solu- porphyrins were extracted as described in the Materials porphyrin that the radioactivity and Methods section and assayed for zinc protoporphyrin tion of zinc chloride supplied is not due to 65Zn2+ (O, in the presence ofether; *, in the absence of ether) and ions. radioactivity (A, in the presence of ether; A, in the absence In an attempt to determine how ether increases of ether). the rate of the reaction the various properties of 39-2 612 A. NEUBERGER AND G. H. TAIT 1964 the enzyme were retested in the presence of ether. porphyrins were unlabelled. Also, if the reaction The heat of activation was 14800 kcal./mole, as it with protoporphyrin, [65Zn]zinc chloride and chro- was in the absence of ether, and the Km values for matophores, in the presence of ether, was allowed the substrates were virtually unchanged. The pH to proceed until about 80 % of the porphyrin was optimum in tris buffer was the same but the shape present as zinc protoporphyrin and then unlabelled of the curve was different, the activity at pH 7 Zn2+ ions were added and the incubation was con- being higher compared with that at pH 8 than it tinued, the molar radioactivity of the isolated zinc was without ether (cf. Fig. 3). Omission of tris protoporphyrin was the same as that obtained buffer had no effect on the enzyme activity. This before adding unlabelled Zn2+ ions. Similarly, if the last point is of importance with regard to the first incubation was with unlabelled Zn2+ ions and specificity for different porphyrins which will be subsequently 65Zn2+ ions were added, only a small discussed below. As in the absence of ether, no amount of radioactivity was found in the isolated other metal ions could replace Zn2+ ions as sub- porphyrins, equivalent in amount to the extra zinc strate. A number of metal salts were tested for protoporphyrin formed during that period of the their ability to inhibit formation of zinc protopor- incubation. These experiments show conclusively phyrin from [65Zn]zinc chloride. At 100 ,m-moles/ that there was no exchange between zinc proto- 0-5 ml. in the presence of 30 tm-moles of zinc porphyrin and Zn2+ ions. chloride the Cu2+ salt inhibited by 45 %, the C02+ As mentioned above, only a small amount of zinc salt by 80 % and the Ca2+ salt by 20 %. protoporphyrin was formed in the absence of After incubations in the presence of ether, in chromatophores and in the presence of tris buffer, which over 80 % of the protoporphyrin was con- and this reaction was not affected by ether. Also, verted into zinc protoporphyrin (as indicated by the enzymic formation of zinc protoporphyrin pro- spectrophotometric measurements), the porphyrins ceeded very well in the absence of tris buffer within were purified (see the Materials and Methods the range pH 7 4-9.0. When other porphyrins were section) and the infrared spectrum was measured. tested for their ability to form zinc porphyrins, the This spectrum was identical in every respect with situation was much more complex (Table 2). With that of authentic zinc protoporphyrin prepared by coproporphyrin and haematoporphyrin all that was alkaline hydrolysis ofzinc protoporphyrin dimethyl required to form the zinc complex was to incubate ester, synthesized according to Fischer & Putzer the porphyrin with Zn2+ ions at neutral pH. After (1926). incubations for 2 hr. a large percentage of the por- When synthetic zinc protoporphyrin, magnesium phyrin was converted into zinc porphyrin, as protoporphyrin or haemin was incubated with judged from the peak in the Soret region of the 65Zn2+ ions andchromatophores, theisolatedmetallo- spectrum, the visible spectrum and the radio-

Table 2. Enzymic and non-enzymic formation of zinc porphyrins Porphyrins ( 50 ,um-moles), 65ZnC12 (30 um-moles; 0-02 pc/mole; 429 counts/min./pm-mole at infinite thinness) and where stated tris buffer, pH 8-4 (100 Zmoles), and chromatophores (1-5 mg. of protein), in a total volume of 0 5 ml. (with the addition of 0-2 ml. of ether where stated), were incubated at 370 for 2 hr. The porphyrins were isolated and the final ether solution containing them was assayed spectrophotometrically in the Soret and visible regions of the spectrum and the radioactivity was determined. -, Not determined. No ether Substrate Ether A r Not Additions ... None Enzyme Tris Tris Enzyme incubated Protoporphyrin Soret peak 404 412 404 404 404 404 Counts/min./ml. of ether 101 970 101 100 150 22 Deuteroporphyrin Soret peak 397 405 404 404 406 396 Counts/min./ml. of ether 373 1760 1126 1736 Mesoporphyrin Soret peak 395 407 406 396 398 396 Counts/min./ml. of ether 85 1010 879 290 Haematoporphyrin Soret peak 410 410 410 - 396 Counts/min./ml. of ether 804 - 830 - - Coproporphyrin Soret peak 407 407 408 408 - 396 Counts/min./ml. of ether 2010 _ 1950 Vol. 90 ZINC-PROTOPORPHYRIN CHELATASE 613 activity in the purified porphyrin fraction when the activity in chromatophores from both sets of pig- reaction was carried out with 65Zn2+ ions. As with mented organisms was also tested. The enzyme zinc protoporphyrin the maxima of absorption in activity in protoplasts and chromatophores from the Soret region of other zinc porphyrins are about both types of pigmented organisms was the same, 10 m, higher than those of the corresponding por- but that inprotoplasts from unpigmented organisms phyrins. Tris buffer, chromatophores and ether had was markedly lower. no effect on these reactions. An incubation was The chromatophore preparations used in the ex- carried out with coproporphyrin and Zn2+ ions in periments described in this paper are not pure and the presence of ether, and as judged from the visible consist of at least two types of particles termed spectrum most of the porphyrin was in the form of 'heavy' and 'light' by Cohen-Bazire & Kunisawa the zinc complex. The infrared spectrum showed a (1960). 'Pure' chromatophores have been prepared peak at 5-9p characteristic of un-ionized carboxyl by Dr A. Gorchein in this Laboratory by sucrose- groups but there were no peaks at 6-4 and 7-1 i, gradient centrifugation of crude 'chromatophores'. which are characteristic of ionized carboxyl groups The 'light' pigmented particles accumulate in (Schwartz, Berg, Bossenmaier & Dinsmore, 1960), 0-66M-sucrose and consist of spheres 400-800 A in indicating that the material is a zinc-copropor- diameter (A. Gorchein, unpublished work; cf. phyrin complex and not a zinc coproporphyrin salt Schachman, Pardee & Stanier, 1952). The activity (Schwartz et al. 1960). With deuteroporphyrin and of zinc-protoporphyrin chelatase in these particles mesoporphyrin only a small amount of zinc com- was as high, on a protein basis, as that of 'crude' plex was formed in the absence of tris and chroma- chromatophores. tophores, andthe reaction couldbe speeded byadding either of these components. With deuteroporphyrin Ferrochelatase in chromatophores ether had no effect on the enzymic or non-enzymic reaction, but with mesoporphyrin virtually no zinc Ferrochelatase activity was detected in chroma- mesoporphyrin was formed in the absence of ether. tophores and was 6-4 pm-moles/mg. of protein/hr. There was no activity in the absence of ferrous sulphate or of ascorbic acid or GSH. Chromatophores Zinc-protoporphyrin-chetatase activity in Rhodo- previously heated to 550 for 5 min. lost about 80 % pseudomonas spheroides grown under different of their activity. The assay was repeated in the conditions presence of 0-2 ml. of ether, which increased the Activity of zinc-protoporphyrin chelatase was rate of the reaction from 6-4 to 19-7 ,um-moles/mg. examined (Table 3) in lysed protoplasts from of protein/hr. With mesoporphyrin, deuteropor- organisms grown at 'high' oxygen partial pressure phyrin andhaematoporphyrin there was a small dis- in the dark (unpigmented), at a 'low' oxygen appearance of porphyrin in the absence of chroma- partial pressure in the dark (pigmented) and tophores but a considerably greater disappearance anaerobically in the light (pigmented). The enzyme in their presence. With these porphyrins the en-

Table 3. Activity of zinc-protoporphyrin chelatase in fractions of Rhodopseudomonas spheroides grown under different conditions Organisms were grown, and lysed protoplasts and 'crude' chromatophores were prepared as described in the Materials and Methods section. Pure chromatophores were prepared by Dr A. Gorchein by sucrose-gradient centrifugation (cf. Bull & Lascelles, 1963). Enzymic assay: tris, pH 8-4 (150 ,umoles), protoporphyrin (50 tom-moles), ZnCl2 (30 ,um-moles) and enzyme source (0-5-3-0 mg. of protein), in a volume of 0-5 ml. plus 0-2 ml. of ether, were mixed thoroughly and incubated at 370 for 60 min. The porphyrins were isolated and purified and the amount of zinc protoporphyrin formed was estimated as described in the Materials and Methods section. Zinc protoporphyrin (em-moles/mg. Expt. Preparation Growth conditions of protein/hr.) 1 (A) Lysed protoplasts High 02, dark 2-8 Lysed protoplasts Anaerobic, light 7-3 1 (B) Lysed protoplasts Low 02, dark 11-6 Lysed protoplasts Anaerobic, light 13-9

Crude chromatophores Low 02, dark 22-0 Crude chromatophores Anaerobic, light 21-6 2 Crude chromatophores Anaerobic, light 22-2 Pure chromatophores Anaerobic, light 18-0 614 A. NEUBERGER AND G. H. TAIT 1964 Table 4. Comparison of chelatase activities in mitochondria and chromatophores The assay conditions for zinc-protoporphyrin chelatase are those described in Table 3 and those for ferrochelatase are described in the Materials and Methods section. To the 0-5 ml. assay system 0-2 ml. of ether was added where stated. The rates of the reactions were calculated as described in the Materials and Methods section. -, Not determined. Production (,um-moles/mg. of protein/hr.) Zinc-protoporphyrin chelatase Ferrochelatase Enzyme source Without ether With ether Without ether With ether Chromatophores 4-8 22-5 6-4 19-7 Guinea-pig liver mitochondria 1-9 18-3 4-4 10-7 Rabbit-liver mitochondria 2-5 1-8 3-2 Rabbit-kidney mitochondria 2-6 4-0 Rabbit-heart mitochondria 6-3 5-0 9-0

zymic activity was slightly less than with proto- mitochondria from guinea-pig liver and from rabbit porphyrin. Coproporphyrin formed coprohaem non- liver, heart and kidney (Table 4). The activities enzymically and addition of chromatophores did were low in the absence of ether, but as with not speed this reaction. chromatophores they were markedly increased in the presence of ether. Inhibition of zinc-protoporphyrin chelatase in chro- When rabbit-liver mitochondria were preheated matophores by Fe2+ ions and of ferrochelatase by at 550 for 5 min. there was almost complete in- Zn2+ ions activation of ferrochelatase activity. The zinc-pro- Although Fe2+ ions do not inhibit zinc-proto- toporphyrin-chelatase activity was not affected porphyrin chelatase under the assay conditions under these conditions but was reduced to 20 % described in Fig. 1, a marked inhibition is observed after heating at 700 for 5 min. in the presence of 20 .tmoles ofascorbic acid. In the Both enzyme activities could be completely ex- presence of 12 tam-moles of zinc chloride, 15 jim- tracted from guinea-pig liver and rabbit-liver mito- moles of ferrous sulphate inhibited by 50 %. Under chondria by freezing in the presence of Tween 20 these conditions only a small amount of haem was (10 mg./ml.) followed by thawing and centrifuging formed. When the reaction was carried out in the at 36000g for 1 hr. (Labbe & Hubbard, 1961). presence of different amounts of Zn2+ and Fe2+ ions After the preparation from rabbit-liver mitochon- the inhibition was competitive, and Ki for ferrous dria had been stored for 2 weeks at -20° all the sulphate was 50 pm. The inhibition was much less zinc-protoporphyrin-chelatase activity had dis- marked in the presence of ether: for instance, appeared, whereas the ferrochelatase was still 100 ,im-moles of ferrous sulphate in the presence of almost as active as before. This finding, together 10 tam-moles of zinc chloride inhibited zinc proto- with the difference in heat inactivation, is evidence porphyrin formation by 85 % in the absence of that in rabbit-liver mitochondria two different en- ether and in zymes catalyse the two reactions, i.e. incorporation by only 15 % its presence. When ferro- of chelatase activity was assayed in the presence of Zn2+ and Fe2+ ions into protoporphyrin. ether Zn2+ ions were inhibitory. Zinc chloride (20 ,im-moles) inhibited the reaction by about 30 % DISCUSSION and 100 in ,um-moles by about 63 % the presence of It is amounts of ferrous sulphate varying between 20 proposed that the incorporation of zinc into and 100 ,im-moles. Thus inhibition by Zn2+ ions is protoporphyrin, which is demonstrated in the pre- non-competitive. sent experiments, like that of the incorporation of ferrous iron, is an enzyme-catalysed reaction. The evidence is that the activity is heat-labile, that it Zinc-protoporphyrin-chelatase and ferrochelatase has a heat of activation of 14800 cal./mole, which activities in mitochondria is similar to that for some other enzyme reactions, Tests for zinc-protoporphyrin-chelatase activity that it has a pH optimum, and that it is specific were carried out with tissues which are known to for the metal. contain ferrochelatase, although previous work Although the formation of zinc protoporphyrin with rat-liver mitochondria (Labbe & Hubbard, and haem is catalysed by the same extracts, our 1961) and duck blood (Oyama et al. 1961) had in- evidence indicates that two different enzymes are dicated little or no zinc-protoporphyrin-chelatase responsible. activity. Both zinc-protoporphyrin-chelatase and (1) In both chromatophores and mitochondria ferrochelatase activities could be detected in zinc protoporphyrin is formed in the presence and Vol. 90 ZINC-PROTOPORPHYRIN CHELATASE 615 the absence of ascorbic acid, whereas haem is Fantl (1963) have shown that a number of hydro- formed only in its presence. philic cations can be transferred from an aqueous (2) GSH inhibits the biosynthesis of zinc proto- to a lipophilic phase by shaking a solution of the porphyrin but not that of haem. cation salt with phosphatidic acid and ether. (3) Extracts prepared from mitochondria by The finding of an enzyme catalysing the forma- treatment with Tween 20 lose their zinc-protopor- tion of zinc protoporphyrin raises the problems of phyrin-chelatase activity but not their ferrochelat- its function and of the role of the product of the ase activity after storage at -20° for about reaction. Although much work has been done on 14 days. the metabolism of zinc in animals, plants and (4) With chromatophores Fe2+ ions competitively micro-organisms and the importance of zinc as a inhibit zinc protoporphyrin formation, whereas component of a number of enzymes is known (cf. Zn2+ ions are a non-competitive inhibitor of ferro- Vallee, 1959), there is very little evidence for any chelatase. connexion between the metabolism of zinc and that Final proof for the existence of two enzymes will, of porphyrin. Zinc uroporphyrin and zinc copro- however, have to await further purification. porphyrin have been identified in the urine of The marked activity of zinc-protoporphyrin patients with acute porphyria (Watson & Schwartz, chelatase demonstrated, in the present study, in a 1941) and in culture filtrates of Corynebacterium number of animal tissues is in contrast with the diphtheriae (Coulter & Stone, 1938). Such com- rather low incorporation of zinc into protopor- plexes can, however, be formed non-enzymically phyrin reported by previous workers. These differ- (Heikel, Lockwood & Rimington, 1958; and the ences may be due to the fact that they used Tween present paper). Zinc deficiency in plants is known in the preparation of the enzyme and GSH or to interfere with chlorophyll synthesis (Vallee, cysteine in its assay. These agents have been found, 1959), but the exact role of zinc has not been in the present work, to destroy or inhibit zinc- elucidated. Zinc has also been implicated in photo- protoporphyrin-chelatase activity respectively. synthesis in Chlorella (Schwartz, 1956). Feeding Moreover none of the previous workers incor- excess of zinc to rats resulted in retarded growth porated ether into their incubation mixtures. and also gave rise to a microcytic hypochromic The enzyme in chromatophores is specific for anaemia (Smith & Larson, 1946). A mixture of Zn2+ ions; no other metalloporphyrins were formed iron, copper and cobalt raised the haemoglobin to when the assay was performed in the absence of normal amounts. The anaemia could be caused by ascorbic acid. Under these conditions only C02+, zinc inhibiting the biosynthesis of haem as has been Mn2+ and Cu2+ ions, and to a lesser extent Ca2+ shown in this work with chromatophores. Dance- ions, were inhibitory. wicz & Malinowska (1958) reported that zinc proto- Many porphyrins rapidly form complexes with porphyrin dimethyl ester, but not the methyl esters zinc in the absence of an enzyme. With mesopor- of protoporphyrin, copper protoporphyrin or cobalt phyrin and Zn2+ ions both tris buffer and ether are protoporphyrin, at concentrations ofbetween 3 and required, with deuteroporphyrin only tris buffer is 100 ,um-moles/ml., caused a 19 % increase in necessary, and with coproporphyrin and haemato- oxygen consumption by rat-liver homogenates. porphyrin neither tris nor ether needs to be present These results do not give much insight into the for complex-formation. Zinc mesoporphyrin and possible functions of zinc protoporphyrin. Exten- zinc deuteroporphyrin can also be formed in the sive experiments carried out in this Laboratory with absence of tris buffer but in the presence of chroma- cell-free extracts have failed to provide any evidence tophores. It is possible that the different behaviour that, in R. 8pheroides, zinc protoporphyrin is an of protoporphyrin as compared with that of the intermediate in the biosynthesis of bacterio- other porphyrins is connected with a more extensive chlorophyll. resonance that can occur in the former porphyrin, SUMMARY which contains two vinyl groups. The reason for the stimulation of zinc-protopor- 1. An enzyme, zinc-protoporphyrin chelatase, phyrin-chelatase activity by ether may be that the has been detected in chromatophores from Rhodo- enzyme is a lipoprotein or the enzyme may be p8eudomona8 8pheroides. It catalyses the formation buried in the chromatophores and mitochondria, of zinc protoporphyrin from zinc chloride and proto- both of which have a high content of lipids. The porphyrin. involvement of lipid is also shown by the finding 2. Zinc-protoporphyrin chelatases are also pre- that the enzyme, along with the ferrochelatase, can sent in mitochondria from guinea-pig liver and be extracted from mitochondria by treatment with from rabbit liver, heart and kidney. Tween 20. Ether may also be acting to transfer 3. The rate of the reaction is about five times as protoporphyrin and Zn2+ ions into a lipophilic high in a water-ether emulsion as in an aqueous phase. Protoporphyrin is ether-soluble and Ward & medium. Methanol and Tween 80 also stimulate, 616 A. NEUBERGER AND G. H. TAIT 1964 but to a smaller extent. The possible significance of Gibson, K. D., Neuberger, A. & Tait, G. H. (1963). these findings is discussed. Biochem. J. 88, 325. 4. The enzyme from R. spheroides is specific for Granick, S. (1961). J. biol. Chem. 236, 1168. Zn2+ ions. Of the salts tested as inhibitors only Heikel, T., Lockwood, W. H. & Rimington, C. (1958). those with Cu2+, Nature, Lond., 182, 313. Co2+, Mn2+ and Ca2+ ions are Hogeboom, G. H. (1955). In Method8 in Enzymology, vol. 1, active. p. 16. Ed. by Colowick, S. P. & Kaplan, N. 0. New 5. When ascorbic acid is present in the assay York: Academic Press Inc. mixture ferrochelatase activity can be demon- Karunairatnum, M. C., Spizizen, J. & Gest, H. (1958). strated in all the above extracts. Under these con- Biochim. biophy8. Acta, 29, 649. ditions Fe2+ ions competitively inhibit zinc proto- Labbe, R. F. & Hubbard, N. (1961). Biochim. biophy8. porphyrin formation and Zn2+ ions are a non-com- Ada, 52, 130. petitive inhibitor of haem formation. Lascelles, J. (1956). Biochem. J. 62, 78. 6. The specificity of the zinc-protoporphyrin Lascelles, J. (1959). Biochem. J. 72, 508. chelatase towards other has Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, porphyrins been tested. R. J. (1951). J. biol. Chem. 193, 265. Unlike protoporphyrin some porphyrins readily Oyama, H., Sugita, Y., Yoneyama, Y. & Yoshikaay, H. form a zinc complex in the absence of enzyme. (1961). Biochim. biophys. Acta, 47, 413. Porra, R. J. & Jones, 0. T. G. (1963a). Biochem. J. 87, 181. REFERENCES Porra, R. J. & Jones, 0. T. G. (1963b). Biochem. J. 87, 186. Schachman, H. K., Pardee, A. B. & Stanier, R. Y. (1952). Bray, G. A. (1960). Analyt. Biochem. 1, 279. Arch. Biochem. Biophye. 38, 245. Bull, M. J. & Lascelles, J. (1963). Biochem. J. 87, 15. Schwartz, H. C., Hill, R. L., Cartwright, G. E. & Wintrobe, Cohen-Bazire, G. & Kunisawa, R. (1960). Proc. nat. Acad. M. M. (1959). Fed. Proc. 18, 545. Sci., Wash., 40, 1543. Schwartz, M. (1956). Biochim. biophy8. Ada, 22, 463. Coulter, C. B. & Stone, F. M. (1938). Proc. Soc. exp. Biol., Schwartz, S., Berg, M. H., Bossenmaier, I. & Dinsmore, H. N. Y., 38, 423. (1960). Meth. biochem. Anal. 8, 221. Dancewicz, A. M. & Malinowska, T. (1958). Postepy Hig. Smith, S. E. & Larson, E. J. (1946). J. biol. Chem. 163, 29. Med. d68w. 12, 447; cited from Chem. Abstr. (1962), 56, Vallee, B. L. (1959). Physiol. Rev. 39, 443. 4046h. Ward, H. A. & Fantl, P. (1963). Arch. Biochem. Biophys. Dresel, E. I. B. & Falk, J. E. (1956). Biochem. J. 63, 388. 100, 338. Fischer, H. & Putzer, B. (1926). Hoppe-Seyl. Z. 154, 62. Watson, C. J. & Schwartz, S. (1941). J. clin. Invest. 20, 440. Gibson, K. D., Neuberger, A. & Tait, G. H. (1962). Yoneyama, Y., Oyama, H., Sugita, Y. & Yoshikaya, H. Biochem. J. 83, 539. (1962). Biochim. biophys. Acta, 62, 261.

Biochem. J. (1964) 90, 616 The Mechanism of Carbohydrase Action 10. ENZYMIC SYNTHESIS AND PROPERTIES OF 6-oc-MALTOSYLGLUCOSE*

BY D. FRENCH,t PAMELA M. TAYLOR AND W. J. VHELAN The Lister Institute of Preventive Medicine, Chelsea Bridge Road, London, S.W. 1 (Received 4 September 1963) The main polymeric linkages of amylopectin and 1958). Isomaltotriose would also be formed if two glycogen are the chain-forming oc-(1--4)-bond and (1-+6)-bonds were in juxtaposition and was indeed the branch-forming a-(1--6)-bond. Partial acid isolated from glycogen by Wolfrom & Thompson hydrolysis of the polysaccharides yields glucose, (1957). The question therefore arises whether the maltose and isomaltose. Of the expected four trisaccharides not detected are not formed, or are trisaccharides [maltotriose, 4-oc-isomaltosylglucose particularly labile to acid and so do not accumu- (panose), 6-oc-maltosylglucose and 4,6-di-ca-glu- late. A partial solution to the problem would be cosylglucose], only maltotriose and panose have obtained by examination of the trisaccharides in been isolated despite a thorough search for the question. Summer & French (1956) indicated that isomers of panose (Edwards, 1955; see also Whelan, these compounds, 6-oc-maltosylglucose and 4,6-di- * Part 9: Allen & Whelan (1963). oc-glucosylglucose, could be synthesized by the t Present address: Department of Biochemistry and coupling reaction between cyclomaltohexaose Biophysics, Iowa State University, Ames, Iowa, U.S.A. (Schardinger c-dextrin) and isomaltose, catalysed